The primary objective of this proposal is to compare the effect of two opioid dependence treatment delivery models on improving the HIV care engagement and outcomes in Vietnam. We partner with several key institutions in Vietnam to propose a multi-site, randomized trial of HIV clinic- based buprenorphine/naloxone (BUP/NX) (n=225) versus referral to methadone maintenance treatment (MMT) (n=225) for treatment of opioid dependence in HIV-infected patients new to care in 5 HIV clinics in Hanoi and Hai Phong-epicenters of the Vietnam HIV epidemic. HIV providers will be trained to manage opioid dependence using flexible BUP/NX dosing and take-out schedules. All subjects will receive clinic-based counseling and HIV care. Self-reported drug use, urine drug screens (UDS), and measures of HIV care engagement and outcomes will be assessed through 12 months follow-up. Qualitative interviews in years 1 through 4 will identify lessons-learned to inform national BUP/NX guidelines for scale-up. We will assess the effect of clinic-based BUP/NX versus MMT referral on HIV viral suppression and treatment engagement (Aim 1) and opioid use (Aim 2) at 12 months. We hypothesize that subjects randomized to clinic-based BUP/NX will have greater HIV viral suppression (HIV-1 RNA pcr < 200 copies/mL), antiretroviral therapy initiation, HIV care retention, and decreased opioid use at 12 months compared with MMT referral. We will test these hypotheses using mixed logistic regression models. Annual qualitative interviews with HIV clinic providers, administrators, and patients will document formative implementation strategies, challenges, and best practices to inform national scale-up of clinic-based BUP/NX. The proposed work is the first large-scale randomized trial of HIV clinic-based BUP/NX versus referral for MMT in real-world HIV clinic settings. A successful study is likely to change HIV treatment practices in Vietnam for patients with co-occurring opioid dependence and inform practices in the United States and worldwide.